Category: Preclinical Development
Purpose: Organic anion transporters (OATs) exhibit broad tissue distribution, including intestine, kidney, liver, choroid plexus and brain capillaries. They play a pivotal role in the absorption, tissue distribution, and excretion of a diverse array of substances, including drugs, toxins, environmental pollutants (xenobiotics) and neurotransmitters. OAT substrates exhibit a wide variety of structural and physiochemical properties, however current knowledge regarding the nature of biochemical interactions between substrate and transporter is sparse. Therefore, identifying which amino acid residues are critical for transporter-substrate interactions is of central importance to understanding and predicting how OATs recognize substrate molecules. The purpose of the present work was to identify amino acid residues mediating interactions within the substrate-binding pocket of human organic anion transporter 1 (hOAT1), which will support breakthroughs in understanding drug absorption, drug-drug interaction, and compound excretion.
Methods: In the absence of any crystalized hOAT, we developed a homology model of hOAT1 using the Piriformospora indica high-affinity phosphate transporter (PiPT) as template. A putative binding pocket for the prototypical OAT substrate para-aminohippurate (PAH) was located and several amino acid residues potentially involved in hOAT1-PAH interactions were identified. Conservative and non-conservative amino acid substitutions were introduced at each position via site-directed mutagenesis using synthetic oligonucleotide primers (QuikChange Lightning Kit). Mutant constructs were confirmed by DNA sequencing and transfected into Chinese hamster ovary (CHO) cells using cationic lipid-based transfection followed by antibiotic selection (G418). Initial screening assays to determine mutant transporter activity were performed with 5µM PAH (0.25 µCi/mL [3H]PAH added as tracer) ± 500 µM probenecid for 10 minutes. Saturation analysis on transport active mutants was conducted using increasing concentrations of PAH (1µM – 200µM; 0.25 µCi/mL [3H]PAH) for 1 minute. Experiments were repeated a minimum of three times. All screening and saturation data are reported as mean ± SD. Final Km estimates are mean ± SE.
Results: PAH docking studies revealed five amino acid positions potentially involved in hOAT1-PAH binding: Arg15, Ile19, Tyr230, Asn439 and Arg466. Initial activity screening experiments indicated that three conservative mutants, Arg15Lys, Ile19Leu, and Tyr230Phe, were still capable of PAH transport, while the hOAT1-Asn439Gln mutant failed to exhibit PAH transport activity. The conservative mutation at position 466, Arg466Lys, was not isolated due to technical reasons. Non-conservative substitution of Ala at all five positions resulted in complete loss of transport activity. Kinetic analysis of the three functional mutants yielded Km values of 16.1 ± 1.0 µM (Arg15Lys), 26.8 ± 1.9 µM (Ile19Leu), and 20.1 ± 2.5 µM (Tyr230Phe), with hOAT1 wild-type being 26.1 ± 2.6 µM, consistent with values reported in the literature. Thus, the conservative Arg15Lys and Tyr230Phe mutants trended toward higher affinity, while Ile19Leu trended toward lower affinity, when compared to hOAT1 wild-type. However, analysis using 1 way ANOVA did not reveal any statistically significant (p > 0.05) changes in PAH affinity.
Conclusion: Preliminary data indicate that in every case non-conservative substitution led to loss of transport activity. Conservative substitution at positions 15, 19 and 230 did not significantly alter transporter affinity for PAH. This suggests some tolerability in sequence changes in hOAT1 without loss of transporter function. Loss of transport activity for the conservative substitution at position 439 may indicate that physiochemical properties at position 439 are less flexible, i.e., this residue may play a major role in PAH binding to hOAT1. Finally, current data are consistent with (1) the PiPT based homology model of hOAT1 at least partially reflecting true hOAT1 3-D structure and (2) that the putative binding pocket investigated contains residues from multiple transmembrane domain regions.